The overall goal of this project is to develop a test for measuring the concentration of unbound unconjugated bilirubin (Bf) to more accurately diagnose than current methods, bilirubin-induced neurotoxicity in newborns. Although more than 70% of newborns have elevated blood bilirubin levels, for most infants this hyperbilirubinemia is benign. However, for some infants bilirubin effects can be neurotoxic, resulting in deficits ranging from reversible hearing defects to the more severe and possibly fatal neurological sequelae of kernicterus. In particular very poor outcomes occur in extremely low birth weight (ELBW) infants where approximately 25-30% die and an additional 25-30% have severe neurologic deficits (e.g. cerebral palsy). This is presumed to be due to bilirubin neurotoxicity. Bilirubin neurotoxicity in neonates is currently assessed by monitoring total serum bilirubin (TSB) and high levels cause phototherapy and/or exchange transfusion to be initiated. However, recent studies and in particular a new study of our own in ELBW infants demonstrates that using TSB will not detect the underlying bilirubin toxicity (high false negativity) and using phototherap based on TSB levels will not improve the very poor outcomes in these infants. Conversely, studies in term infants presenting with extremely high TSB levels (e 25 mg/dL), the threshold for exchange transfusion, have recently found that the exchange transfusion (itself associated with ~ 2% mortality) may have been unnecessary in as many as 80% of these infants (high false positivity). This implies that the percent of infants unnecessarily treated with phototherapy because of high TSB may be even larger because the threshold for phototherapy treatment is lower. The reason for the failure of TSB as a diagnostic is that bilirubin is highly insoluble and n blood the vast majority is bound to albumin, normally only about 10-4 is in the unbound or free state (Bf). However, only the Bf fraction of TSB can cross the blood brain barrier and cause kernicterus. TSB cannot account for the many factors that change bilirubin-albumin binding and therefore Bf. What is needed to improve outcomes and reduce unnecessary therapy, is an accurate and simple to use method for measuring Bf. The specific Aims of the project are, in Phase I, to develop a simple disposable cartridge containing our recently developed fluorescently labeled mutant fatty acid binding proteins (FABPs) that allow accurate quantitation of Bf by a change in fluorescence upon binding Bf. These probes will use near infrared (NIR) fluorescence and thereby allowing measurements of Bf in 5 ?L of undiluted whole blood applied to the cartridge. In Phase II we will develop a ratio fluorescence instrument that is optimized for measuring Bf on sample cartridge and will display the Bf concentration in less than 1 minute.